Jiahuan Zhou

Peking University, Nanjing Tech University

Papers

Total Citations

108

H-Index

About

Jiahuan Zhou is a leading researcher in computer vision and robotics, with a primary focus on person re-identification (Re-ID) and human-robot interaction. His most impactful contribution is the development of **DCR-ReID (Deep Component Reconstruction)**, a groundbreaking framework for cloth-changing person re-identification (CC-ReID). This work addresses the critical challenge of recognizing individuals when their appearance changes due to different clothing over time—a problem that plagues long-term surveillance and forensic applications. With **91 citations**, DCR-ReID has become a cornerstone in the field, enabling robust identification by reconstructing identity-discriminative components that are invariant to clothing. Beyond Re-ID, Zhou has advanced object detection and tracking for human-robot interaction, notably implementing YOLO networks on the NAO robot to enhance its "I See You" function for object handover tasks. He has also explored the theoretical underpinnings of deep learning, analyzing how different pooling strategies in CNNs affect invariance to position and deformation—work that informs more robust visual recognition systems for robotics. Zhou’s research bridges fundamental computer vision challenges with practical robotic applications, making him a key figure in developing AI systems that can perceive and interact with humans reliably in dynamic, real-world environments.